Przemyslaw Stefan Rokita

Generating depth of-field effects in virtual reality applications

Computer-generated images without depth-of-field effects always look surreal. A new technique and algorithm are presented which solve the problem of simulating these effects in VR and other real-time applications. The adaptive convolution approach can generate depth-of-field effects in real time using commercially available hardware filters. The picture-processing algorithm proposed can be easily implemented in a visualization pipeline. The final effect can be obtained by iterative filtering with a single hardware convolution filter or with identical 3/spl times/3 filters placed as pipeline stages. The speed of this approach makes it suitable for generating images in real-time applications, particularly virtual reality systems. As a postprocessing technique, it also has the advantage of being easily implemented as a final stage of the visualization pipeline in an existing rendering system.

Fast generation of depth of field effects in computer graphics

Abstract Depth of field and, in some circumstances, uniform defocusing are very important effects that should be taken into account for generating more realistic images, closely approximating a scene imaged by a real camera or seen by the human eye under poor visual conditions. Some algorithms for generation of such effects have already been given by other authors, but they are computationally expensive. This paper proposes a new technique for generating depth of field and defocus effects by adaptive smoothing filtering. Proposed algorithm is fast and suitable for real time applications, such as, for example, visual simulators.

Perception-based fast rendering and antialiasing of walkthrough sequences

We consider accelerated rendering of high quality walkthrough animation sequences along predefined paths. To improve rendering performance, we use a combination of a hybrid ray tracing and image-based rendering (IBR) technique and a novel perception-based antialiasing technique. In our rendering solution, we derive as many pixels as possible using inexpensive IBR techniques without affecting the animation quality. A perception-based spatiotemporal animation quality metric (AQM) is used to automatically guide such a hybrid rendering. The image flow (IF) obtained as a byproduct of the IBR computation is an integral part of the AQM. The final animation quality is enhanced by an efficient spatiotemporal antialiasing which utilizes the IF to perform a motion-compensated filtering. The filter parameters have been tuned using the AQM predictions of animation quality as perceived by the human observer. These parameters adapt locally to the visual pattern velocity.

A model for rendering high intensity lights

Abstract In photo-realistic image rendering, light is the most important factor, which is determining if the generated image of a rendered scene will be perceived by the observer as the image of a real scene. Actual global lighting models are based on reflection and diffusion of light on surfaces of the modeled scene. This approach cannot render high intensity effects and glare effects, which are important not only for the artistic impression of the observer, but they are crucial in drive simulators. This paper proposes a new model for rendering high intensity lights, blooming, and glare effects. This model is based on the human eye structure and its physiology. An efficient algorithm for rendering those phenomena is also given.

Manipulating refractive and reflective binocular disparity

Presenting stereoscopic content on 3D displays is a challenging task, usually requiring manual adjustments. A number of techniques have been developed to aid this process, but they account for binocular disparity of surfaces that are diffuse and opaque only. However, combinations of transparent as well as specular materials are common in the real and virtual worlds, and pose a significant problem. For example, excessive disparities can be created which cannot be fused by the observer. Also, multiple stereo interpretations become possible, e. g., for glass, that both reflects and refracts, which may confuse the observer and result in poor 3D experience. In this work, we propose an efficient method for analyzing and controlling disparities in computer‐generated images of such scenes where surface positions and a layer decomposition are available. Instead of assuming a single per‐pixel disparity value, we estimate all possibly perceived disparities at each image location. Based on this representation, we define an optimization to find the best per‐pixel camera parameters, assuring that all disparities can be easily fused by a human. A preliminary perceptual study indicates, that our approach combines comfortable viewing with realistic depiction of typical specular scenes.

Spatio-temporal photon density estimation using bilateral filtering

Photon tracing and density estimation are well established techniques in global illumination computation and rendering of high-quality animation sequences. Using traditional density estimation techniques it is difficult to remove stochastic noise inherent for photon-based methods while avoiding overblurring lighting details. In this paper we investigate the use of bilateral filtering for lighting reconstruction based on the local density of photon hit points. Bilateral filtering is applied in spatio-temporal domain and provides control over the level-of-details in reconstructed lighting. All changes of lighting below this level are treated as stochastic noise and are suppressed. Bilateral filtering proves to be efficient in preserving sharp features in lighting which is in particular important for high-quality caustic reconstruction. Also, flickering between subsequent animation frames is substantially reduced due to extending bilateral filtering into temporal domain

GPU-Based Hierarchical Texture Decompression

We propose a hierarchical texture compression algorithm for real-time decompression on the GPU. Our algorithm is characterized by low computational complexity, random access and hierarchical structure which allow us to access first three levels of encoded mip-map pyramid. The hierarchical texture compression algorithm HiTCg is based on block-wise approach, where each block is subject to local fractal transform.

Efficient multi-image correspondences for on-line light field video processing

Light field videos express the entire visual information of an animated scene, but their shear size typically makes capture, processing and display an off‐line process, i. e., time between initial capture and final display is far from real‐time. In this paper we propose a solution for one of the key bottlenecks in such a processing pipeline, which is a reliable depth reconstruction possibly for many views. This is enabled by a novel correspondence algorithm converting the video streams from a sparse array of off‐the‐shelf cameras into an array of animated depth maps. The algorithm is based on a generalization of the classic multi‐resolution Lucas‐Kanade correspondence algorithm from a pair of images to an entire array. Special inter‐image confidence consolidation allows recovery from unreliable matching in some locations and some views. It can be implemented efficiently in massively parallel hardware, allowing for interactive computations. The resulting depth quality as well as the computation performance compares favorably to other state‐of‐the art light field‐to‐depth approaches, as well as stereo matching techniques. Another outcome of this work is a data set of light field videos that are captured with multiple variants of sparse camera arrays.

Compositing computer graphics and real world video sequences

This paper describes new techniques for compositing computer generated images and real world video sequences obtained using a camcorder. The methods described here can be used to produce animations of computer generated images in an accurate real background. Such animations can be applied, for example, in the environmental assessment and pre-evaluation of the visual impact of large scale constructions. The main advantage of the methods proposed here is that in order to track the camera position and merge computer graphics and real images, only the frames from the video sequence are analyzed, without the use of costly sensors usually applied to the camera movement detection. The problems of transition from analog images, captured by a camcorder, to the digital pictures used in computer graphics are also discussed. The proposed algorithms may be directly used in applications of computer graphics such as: multimedia, augmented reality, telepresence, visual simulators, and in all internet/WWW applications requiring those features.

PATTERN MATCHING WITH DIFFERENTIAL VOTING AND MEDIAN TRANSFORMATION DERIVATION Improved Point-Pattern Matching Algorithm for Two-Dimensional Coordinate Lists

We describe an algorithm for matching two-dimensional coordinate lists. Such matching should be immune to translation, rotation, scaling and flipping. We assume that coordinate lists may be only partially overlapping and we allow some random perturbations in the lists. Our goal is to match enough points to be able to derive a coordinates transformation with high confidence. The implementation described here is a part of SkySpy automated sky survey system. The presented algorithm is used to perform matching of stars detected during observations with star catalogues.